It is children living in Africa south of the Sahara that bear the brunt of morbidity and mortality caused by Plasmodium falciparum. Here, traditional methods of vector control have been least effective, and the mitigating role of permethrin-impregnated bednets is threatened by emergence of permethrin resistance in the primary vector, Anopheles gambiae. Effective strategies for monitoring and managing the spread of insecticide resistance will depend upon detailed information about the genetic structure of vector populations. This same information is essential for future genetic control strategies aimed at the replacement of vector by nonvector populations. In this context, it is important to address the hypothesis that different chromosomal forms of An. gambiae are reproductively isolated, using independent molecular markers. It is also a premise of this proposal that the requisite information extends beyond An. gambiae to a second vector of major importance, its sibling species An. arabiensis. The possibility that An. gambiae may be hybridizing productively with An. arabiensis suggests that undesirable traits such as insecticide resistance could be exchanged between species. Past efforts to conclusively demonstrate gene flow between these vectors ahve been complicated by the confounding explanation of recent common ancestry and the contadictory evidence concerning their status as sister taxa within the species complex known as the An. gambia complex. The Y chromosome provides a powerful tool to attack questions at the population level and above, not only because it is nonrecombining and paternally transmitted, but also because it cannot cross barriers since male hybrids are sterile. The main goal of this proposal is the development and application of DNA sequence markers on the Y chromosome (microsatellites, insertion-deletions, and base substitutions) that, in combination with other markers, will help solve three interrelated questions: (1) Are An. gambiae and An. arabiensis sister taxa? (2) How much gene flow is occurring between them, and which genomic regions are susceptible? (3) How much gene flow is occurring between chromosomal forms of An. gambiae?